MINTS-III is a competing continuation application resubmission in response to PA-07-295. It builds upon MINTS-I, one of the first survey studies of the Internet and HIV risk among MSM; and MINTS-II, which developed and tested Sexpulse, an online, HIV prevention intervention for Men who use the Internet to seek Sex with Men (MISM). MINTS-II is the first online RCT of MISM to report acceptable retention over 12 months and short-term efficacy. At the 3-month follow-up, the intervention group reported a significantly greater (20%) reduction in risk behavior than the null control. We also observed a significant reduction in reported risk behavior in the control group, which we hypothesize is the product of panel conditioning. MINTS-III is the logical next step in this promising area of research. Our first aim is to increase the long-term HIV prevention effectiveness of the intervention by strengthening the dosage of Sexpulse. A substantial user-centered design iteration of the intervention will incorporate personalized, tailored feedback on individual risk behavior; integrate research surveys into the intervention; and add follow-up boosters all aimed at increasing the long-term behavioral effectiveness. In Aim 2, we propose two sub-studies to examine the reliability and validity of online measures of sexual risk behavior. Aim 3 tests the long-term risk reduction efficacy of the strengthened intervention, Sexpulse 1.1, and boosters, in an RCT of 1000 high-risk MISM. This RCT design has been enhanced by three levels of recruitment so that in Aim 4, we can study threats to risk measurement, especially panel conditioning. Our final aim tests the long-term protective efficacy of Sexpulse 1.1 (including online boosters) in an RCT on 500 no/lower risk MISM. Upon successful completion of these aims, Sexpulse 1.1 should be a strong and demonstrated-effective intervention that can be strategically positioned for dissemination.